Conference & Journal Papers Database

< Back to search

Choice of licensing method and crashes of young drivers

Anderson, R, Kloeden, C, Hutchinson, T, McLean, A J (Peer reviewed)

Young Drivers

2006

Five years of data (1998-2002) were used to examine whether there was a relationship between the method of driver licensing - Competency Based Training (CBT) or Vehicle On-Road Test (VORT) - and the subsequent crash experience of young drivers, using logistic regression analysis.

The main findings of this study were: Statewide, choosing the VORT was associated with a 6% increase in the odds of having at least one crash in the first 180 days. In one year, if those who chose VORT had an equivalent crash risk to those who chose CBT, there might have been 20 fewer non-casualty and 10 fewer casualty crashes in new drivers’ first six months of driving. The choice of licensing method was less important than the variables: area of residence, sex, age and the period spent on a learner’s licence.

However, choosing VORT rather than CBT could easily be due to factors (amount of travel, personality, social habits) that are also associated with a greater likelihood of crashing. For example, we found that choosing VORT rather than CBT was associated with a 25% increase in the odds that the driver had been involved in a crash as a driver prior to the issue of a P licence. This means that there is a real possibility that the slight increase in the odds of having at least one subsequent crash (noted in 1. above) is not due primarily to any characteristic of the VORT test itself but rather something about the drivers who chose to take the VORT.We therefore found no clear evidence that any differences between the VORT and CBT methods of licensing are related to subsequent crash experience.In separate analyses, we found no evidence that the choice of examiner for the VORT, or the instructor for the CBT test, has any significant influence on subsequent crash outcome. Keywords: Young drivers, risk factors for crash involvement, licensing methods, driving instructor, logistic regression, data linkage.